Data Quality Assessment
Description
Analyzes data profiles, schema patterns, and existing data quality issues to automatically recommend technical data quality rules that improve accuracy, consistency, completeness, compliance, and reliability.
Details
This recipe helps teams assess the health of all queryable tables in the workspace by profiling structure, inferring column behavior, identifying data quality issues, and turning those findings into visual outputs that support prioritization, governance, and remediation planning.
Business value
- Creates a reusable data quality baseline across all eligible workspace tables.
- Highlights nulls, duplicates, non-unique identifiers, constant values, long text anomalies, and possible PII patterns.
- Transforms raw profiling results into summaries and charts that support governance and operational decision-making.
- Helps teams focus remediation effort on the tables, issue types, and severities with the highest impact.
Who benefits
- Data stewards responsible for monitoring quality and enforcing standards.
- Data engineers validating source readiness for downstream pipelines and analytics.
- Analytics and BI teams that need trusted datasets for reporting and dashboarding.
- Governance and compliance stakeholders tracking sensitive data patterns and remediation priorities.
What you receive
How the recipe works (guided flow)
Step 1 — Identifying Dq Issues By Profiling the Datasets
The recipe scans all valid workspace tables, profiles schema and content, infers data types and business roles, detects patterns, computes statistics, and writes structured profiling and DQ result tables for downstream use.
Step 2 — Issue severity and remediation
The recipe reads detailed DQ issue outputs to visualize which tables have the highest issue volume, how those issues break down by severity, and which remediation themes appear most often.
Step 3 — Completeness vs duplicates
The recipe compares completeness percentages and duplicate row counts across tables to make quality gaps easier to spot and prioritize.
Step 4 — Issue type distribution
The recipe aggregates issue categories across the workspace and visualizes the dominant data quality problem types affecting the datasets.
Sample insights (illustrative)
| Insight Category | What the recipe discovered | Business Impact |
|---|---|---|
| Table risk concentration | A small number of workspace tables account for most detected DQ issues, making them the highest-priority targets for corrective action. | Teams can focus effort where cleanup will reduce the greatest volume of downstream reporting and trust issues. |
| Identifier and completeness problems | Profiling can surface null-heavy columns, non-unique identifier candidates, and duplicate-related issues that weaken key integrity. | Improving these fields strengthens joins, reduces data mismatches, and improves confidence in analytics outputs. |
| Remediation planning | Recommended actions can be grouped by issue type, revealing recurring fixes such as deduplication, null handling, and masking of detected PII patterns. | Business and technical teams gain a clearer remediation roadmap instead of reviewing isolated data defects one by one. |
Make sure the following ingredients are available in your workspace:
- Connected datasets are crawled and profiled
- Columns have distinct counts and top values generated